An official website of the United States government
Abstract Researchers typically rely on fossils from the Family Bovidae to generate African paleoenvironmental reconstructions due to their strict ecological tendencies. Bovids have dominated the southern African fauna for the past four million years and, therefore, dominate the fossil faunal assemblages, especially isolated teeth. Traditionally, researchers reference modern and fossil comparative collections to identify teeth. However, researchers are limited by the specific type and number of bovids at each institution. B.O.V.I.D. (Bovidae Occlusal Visual IDentification) is a repository of images of the occlusal surface of bovid teeth. The dataset currently includes extant bovids from 7 tribes and 20 species (~3900). B.O.V.I.D. contains two scaled images per specimen: a color and a black and white (binarized) image. The database is a useful reference for identifying bovid teeth. The large sample size also allows one to observe the natural variation that exists in each taxa. The binarized images can be used in statistical shape analyses, such as taxonomic classification. B.O.V.I.D. is a valuable supplement to other methods for taxonomically identifying bovid teeth.
more »
« less
African bovid tribe classification using transfer learning and computer vision
Abstract Objective analytical identification methods are still a minority in the praxis of paleobiological sciences. Subjective interpretation of fossils and their modifications remains a nonreplicable expert endeavor. Identification of African bovids is a crucial element in the reconstruction of paleo‐landscapes, ungulate paleoecology, and, eventually, hominin adaptation and ecosystemic reconstruction. Recent analytical efforts drawing on Fourier functional analysis and discrimination methods applied to occlusal surfaces of teeth have provided a highly accurate framework to correctly classify African bovid tribes and taxa. Artificial intelligence tools, like computer vision, have also shown their potential to be objectively more accurate in the identification of taphonomic agency than human experts. For this reason, here we implement some of the most successful computer vision methods, using transfer learning and ensemble analysis, to classify bidimensional images of African bovid teeth and show that 92% of the large testing set of images of African bovid tribes analyzed could be correctly classified. This brings an objective tool to paleoecological interpretation, where bovid identification and paleoecological interpretation can be more confidently carried out.
more »
« less
- Award ID(s):
- 2015374
- PAR ID:
- 10655667
- Publisher / Repository:
- Wiley
- Date Published:
- Journal Name:
- Annals of the New York Academy of Sciences
- Volume:
- 1530
- Issue:
- 1
- ISSN:
- 0077-8923
- Page Range / eLocation ID:
- 152 to 160
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Paleoanthropological and geological field research at Galili, Afar Regional State, Ethiopia was reinitiated in 2016. The site has been established as encompassing sediments ranging between 2.5-4.5± Ma and includes early Australopithecus and perhaps Ardipithecus. While preliminary paleoenvironmental analyses have suggested open woodland to bushland-woodland and shrubland, we present here additional evidence of the diet and functional anatomy of the Early Pliocene bovids from Galili. Fossil bovids have been recognized as a crucial tool in understanding ancient environments and serve as an important ecological indicator. Establishing the habitats of Galili is important for both understanding the environment in which the fauna, including hominins, resided and providing comparison to other contemporaneous sites. Taxonomic identification and metric analyses were performed on bovid teeth (n=93) from the 3.9-4.4 Ma Shabeley Laag and Dhidinley members to reconstruct the paleoenvironment. The sample was dominated by Reduncini, a riparian-associated tribe, and Tragelaphini, a browsing-associated tribe. Hypsodonty index was calculated for each tribe and results suggest the most abundant bovid taxa trend towards the brachydont category. Only three tribes: Alcelaphini, Antilopini, and Hippotragini fell within the mesodont category, and no tribes had hypsodont indices. Mesowear analysis of the sample indicated high occlusal relief and rounded cusp shapes were most common. Taken together, our results indicate a trend in bovid diet towards mixed feeding and browsing at Galili.more » « less
-
Paleoanthropological and geological field research at Galili, Afar Regional State, Ethiopia was reinitiated in 2016. The site has been established as encompassing sediments ranging between 2.5-4.5± Ma and includes early Australopithecus and perhaps Ardipithecus. While preliminary paleoenvironmental analyses have suggested open woodland to bushland-woodland and shrubland, we present here additional evidence of the diet and functional anatomy of the Early Pliocene bovids from Galili. Fossil bovids have been recognized as a crucial tool in understanding ancient environments and serve as an important ecological indicator. Establishing the habitats of Galili is important for both understanding the environment in which the fauna, including hominins, resided and providing comparison to other contemporaneous sites. Taxonomic identification and metric analyses were performed on bovid teeth (n=93) from the 3.9-4.4 Ma Shabeley Laag and Dhidinley members to reconstruct the paleoenvironment. The sample was dominated by Reduncini, a riparian-associated tribe, and Tragelaphini, a browsing-associated tribe. Hypsodonty index was calculated for each tribe and results suggest the most abundant bovid taxa trend towards the brachydont category. Only three tribes: Alcelaphini, Antilopini, and Hippotragini fell within the mesodont category, and no tribes had hypsodont indices. Mesowear analysis of the sample indicated high occlusal relief and rounded cusp shapes were most common. Taken together, our results indicate a trend in bovid diet towards mixed feeding and browsing at Galili.more » « less
-
Paleoanthropological and geological field research at Galili, Afar Regional State, Ethiopia was reinitiated in 2016. The site has been established as encompassing sediments ranging between 2.5-4.5± Ma and includes early Australopithecus and perhaps Ardipithecus . While preliminary paleoenvironmental analyses have suggested open woodland to bushland-wood - land and shrubland, we present here additional evidence of the diet and functional anatomy of the Early Pliocene bovids from Galili. Fossil bovids have been recognized as a crucial tool in understanding ancient environments and serve as an important ecological indicator. Establishing the habitats of Galili is important for both understanding the environment in which the fauna, including hominins, resided and providing comparison to other contemporaneous sites. Taxonomic identification and metric analyses were performed on bovid teeth (n=93) from the 3.9-4.4 Ma Shabeley Laag and Dhidinley members to reconstruct the paleoenvironment. The sample was dominated by Reduncini, a riparian-associated tribe, and Tragelaphini, a browsing-associated tribe. Hypsodonty index was calculated for each tribe and results suggest the most abundant bovid taxa trend towards the brachydont category. Only three tribes: Alcelaphini, Antilopini, and Hippotragini fell within the mesodont category, and no tribes had hypso - dont indices. Mesowear analysis of the sample indicated high occlusal relief and rounded cusp shapes were most common. Taken together, our results indicate a trend in bovid diet towards mixed feeding and browsing at Galili. We thank Prof. H Seidler and his team for allowing access to the fossils they collected.more » « less
-
Abstract Accurate cloud type identification and coverage analysis are crucial in understanding the Earth’s radiative budget. Traditional computer vision methods rely on low-level visual features of clouds for estimating cloud coverage or sky conditions. Several handcrafted approaches have been proposed; however, scope for improvement still exists. Newer deep neural networks (DNNs) have demonstrated superior performance for cloud segmentation and categorization. These methods, however, need expert engineering intervention in the preprocessing steps—in the traditional methods—or human assistance in assigning cloud or clear sky labels to a pixel for training DNNs. Such human mediation imposes considerable time and labor costs. We present the application of a new self-supervised learning approach to autonomously extract relevant features from sky images captured by ground-based cameras, for the classification and segmentation of clouds. We evaluate a joint embedding architecture that uses self-knowledge distillation plus regularization. We use two datasets to demonstrate the network’s ability to classify and segment sky images—one with ∼ 85,000 images collected from our ground-based camera and another with 400 labeled images from the WSISEG database. We find that this approach can discriminate full-sky images based on cloud coverage, diurnal variation, and cloud base height. Furthermore, it semantically segments the cloud areas without labels. The approach shows competitive performance in all tested tasks,suggesting a new alternative for cloud characterization.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1111/nyas.15067
